akrenits said:
With a pack this large and a discharge current of 300A, around 5C, there is a decent margin for the pack to become unbalanced during a single discharge, correct?
Not if you have good, matched cells that you are not using outside their nominal specifications. There's good cells out there that have been mentioned or evne used here on ES that could handle that kind of current, though I have no idea if they meet your weight/volume/capacity needs.
For example, I use "ancient-technology" EIG NMC 20Ah cells that could do 5C, and if you had 3P you could get 300A out of them at their max ratings. If you used 6P you'd be easily within their nominal ratings and quite safe with no worries.
I'm sure there are newer cells that can do better than that, but I don't keep up with the tech. You could find the thread about various battery tech announcements; it's somewhere in this subforum. Sorry I don't have a link (I don't read it, just have seen it).
People also make separate threads about new tech, rather than putting htem there, because they would rather clutter the forum with a billion unfindable threads about stuff than keep useful info all in one place, but you'd have to find those threads individually by reading thru the whole forums since I know of no way to search for them, as all the common terms for them are also common for most other things here on ES.
If you abuse your cells outside their specs, or use them at their absolute maximum ratings a lot, then you can expect the pack to behave abnormally, get hotter, and need more maintenance, which would include cell replacements more often, and you would want to design the pack to accomodate that, and as with many aircraft things, require that anything out of spec (cells that cut out early) be replaced immediately.
It'd be a lot better than relying on a system that could cause these problems, if just a single relay were to fail:
--A series relay failure to close would cause complete power loss, with no possibility of manual recovery in-flight. Unless you have a secondary pack, or use two parallel packs of each half the capacity, each of which can still handle the full current load by itself while staying in it's nominal specs, the craft is lost and crew/etc dead because of a single-point failure.
--A series relay failure of closing when not commanded would cause a dead short across a fully-charged cell and cause the cell to burn, if not fused.
-If the cell is fused it would still suddenly drop the voltage of the pack by that cell, losing some power.
-If the controller then commands more current to make up for the voltage drop so the rotors can keep the speed needed for the commanded flight operation, then the cells are now loaded more heavily. You'd have to use cells that are spec'd for the worst-case situation of losing say, half of your pack, for their normal usage, or risk overheating the pack and possibly causing a fire, or at the least damaging the cells and requiring replacement if the aircraft survives for landing.
Those are just two obvious possible single-point failure scenarios that immediately occur to me; I'm sure there's others.
Obviously the proper flight plan is to never reach below 30% charge to accommodate a huge safety margin for this life/death scenario.
If I were doing this, I'd simply mark the fuel gauges to only even show that 70, or less, percent of the pack is actually even available to start with. So Full to Empty on the cockpit gauge is actually only showing you 100% down to 30%. And never tell the pilot that they even *have* the other 30%, so they can't ever plan to use it, but it will still be there if everything goes sideways up in the air and they just "magically" have more fumes than they expected.
For whatever reason if this is not achieved in-flight, and you are too low in altitude to use primary safety options, the BMS overrides would be an emergency option, trying to eek out 30 more seconds of flight to make a proper landing. The bypassing of individual cells would save the whole pack if I only needed that last 2% charge,
If they're well-balanced you won't have 2%; they'll all die at the same time, so you won't have anything to use. Having an emergency system that *depends on* having something that can't be relied on to exist sounds like a recipe for tragedy.
allowing me 15 seconds more flight. In a more severe scenario, I want the option of bypassing over-discharge to get those last 30 seconds.
For that, you'd want to set the LVC a lot higher, so that much more of the pack remains. Not to prevent pack damage, but to prevent the probability that at some point the cell(s) simply can't provide that emergency power, exactly at the time it would be relied upon to be there, because the pilot "knows" it's there (even though it may not be).
I wouldn't want to be responsible for a design that has a likelihood of killing people like that.
There is no price to be put on safety and redundancy in a system like this.
Completely true, which is why you would not want to design it to rely upon capacity that may not even be there, and instead design the pack to hold much more than is actually needed, or derate the pack's stated capability to leave more margin for aging, abuse, etc.
If you are limited by weight or volume, etc., and cells dont' exist that can give you both the Wh and A needed, simultaneously, then you simply have to call the pack a smaller capacity than it actually is, and keep the remainder as your reserve. The pack will not have the range you want, but that's just the way it would have to be, until weight/volume/etc can be made available for the proper-capability pack, or cells become available to create one that does fit within the constraints.
Otherwise the only safe option I can see is to remove weight elsewhere to make a proper-capability pack possible.
Basically, I want the option, and I never want to use that option. Additionally, these modifications would be demonstrating intelligent, safe electrical design for the project.
Not to me. They represent possible single-point failures that will kill people when they happen. :/
- Under 300A load (4-5C) would the balance be off enough that the primary cell bypass would provide even 10 more seconds of flight? How about on an older, well-loved pack?
No way to know till you actually test it with the specific cells and equipment you are going to use. For an older "well-loved" pack, if it can't provide the cabability *AND* reserve required, it must be replaced immediately, not depended upon to do something it possibly (probably) can't do anymore.
And remember that it is not only the cells that have to deal with this situation.
Relays must all handle the currents needed, and they must not arc or weld contacts (or blow semiconductors if they are solid-state), under any conditions that could possibly ever happen on the aircraft (not just normal flight, but conditions that would be just short of tearing it apart or otherwise killing the occupants anyway).
Controller has to handle the voltage drops as cells cut out of the circuit, so you need a controller that basically has no LVC, and is actually capable of still running the motors even on just one cell.
Remember: the cell-bypass system could cause the entire pack to cut out like a set of christmas lightbulbs wired in series, where one bulb burns out and then shorts across itself so the rest of the string stays lit, but then the other bulbs all have a higher load on them, so any that are already stressed will die, putting a higher load on the others, which are now stressed even further, so more will quickly die, putting much more stress on the remaining ones, until they just all go pop and there's no lights left.
Except in this case that means there's no power left, and you're falling out of the air with no hope of recovering from it.
- With a 32s 80Ah pack, would discharging from 3v to empty (under load 2.7v to completely dead) provide enough power to eek out even 15 more seconds of flight?
Dunno--you'd have to test it with the specific cells, once you know exactly how much power is consumed in such an emergency action. Or talk to the company that makes the cells, to see what they have tested them to be capable of.
I would never depend on that last bit of power being there for such an emergency situation. What happens when it isn't?
Amberwolf: The packaging concern is definitely a consideration. The vibrations too, obviously. Is there no possible work-around for this?
I don't know what you mean--vibration is vibration, it's always there, and turbulence in aircraft can be severe enough to tear one apart, under the wrong conditions, so you have to have relays that are totally vibration-proof. I don't even know if they exist, other than solid-state stuff.
Even the relays used on stuff in regular aircraft control systems occasionally have chatter on them, based on what I saw when I was a final QC test tech at Honeywell CFSG, during burn-in vibration tests. It's not such a huge deal there becuase the chatter might only be once for an instant, but it's not high current and it's not handling all the power for the entire aircraft--yours HAS TO handle that...so any chatter means either loss of all power for that instant, or if it's a series relay that shorts across a cell because of it it could arc and weld under the current that will now flow thru it (which will be many times what the relay normally handles, even if it's only until the cell-fuse blows).
Can you elaborate a bit more on the fusing of the batteries? Not sure I understand.
Each cell could have a fuse instead of a relay to cut it out of the pack.
The fuse would be rated to handle "normal" loads (anything the controller can spike out of the pack) but would blow if the cell were shorted, before the wiring to the cell could be damaged from the heat, or the cell itself would heat enough to cause a fire.
The relay that shorts across the cell would be what blows this fuse and disconnects the cell from the pack.
I'd use fuses even if you use a relay so that if the series relay fails shorted you don't have a fire.
It'd also prevent a fire in the case of a wiring fault (broken insulation shorting to something else).
Could you link to some sources for the kinds of relays and such I'd need? Or elaborate on what these components are referred to in the industry? Trying to get an idea of what weight this specialized system would add. Do they make specialized, low-weight electrics like this for aerospace applications?
You'll have to research that stuff; I don't have any idea what's actually out there right now, other than that the higher current stuff is always bigger and heavier and costs more. Smaller options cost even more. (There may not even be anything in existence that does what you need it to do, so you might have to design it and have it custom-made for you.)
If you put that money into better cells, you'd be a lot better off, with less failure points. Less things to go wrong with nothing between you and hard ground but dozens to thousands of feet of air is a good thing.
Arguments can definitely be made that the budget batteries I am referencing are not a fair comparison, but everything is built on SOME budget :wink:
If you have a system that is life and death dependent on having these batteries have the capabilities you need, then you need to consider increasing the battery budget to ensure they can do it even when aged, and design the system around ensuring the batteries are never put into a situation where they can't do it.
. I've been dreaming about making this since 2007 when I was active here. It's only now that I figured out how to make it using Arduino and relay boards. But the relays are only rated at 30VDC 10A which is great for my boombox packs and a 30 volt ebike. I've been flaky so I won't promise that I'll start sharing pictures again but I will be very motivated to do it if my Bypass BMS build is successful 
